Drug and Alcohol Dependence, 25 (1990)179182
Research issues in assessing addiction treatment efficacy: How cost effective are Alcoholics Anonymous and private treatment centers?
Morristown, New Jersey
Therapies in most areas of health care are subjected to clinical trials, e.g., an ethical drug cannot be approved for sale in the United States without having been demonstrated experimentally to be safe and effective. The ideal of the clinical trial is an experiment in which patients are randomly assigned to two or more treatment and control groups and outcomes are compared. Health care research today in addition increasingly addresses cost effectiveness, i.e., how much the cost of the therapy relative to its outcome compares with the cost and benefits of alternatives.
In the case of alcohol and drug abuse treatments, such research has been conducted (primarily in the case of alcoholism therapy). However, the results of such experiments are not funneled into any regulatory body that can approve a therapy. According to Miller and Hester  in the United States 'present policies. . .[entail] few conditions of accountability for quality or effectiveness'. Rather, treatment practices in the United States are based on historical traditions and folk beliefs that owe more to religion and temperance than to research. Enoch Gordis , the current director of the American National Institute on Alcohol Abuse and Alcoholism (NIAAA), describes the treatment environment in the United States:
In the case of alcoholism, our whole treatment system, with its innumerable therapies, armies of therapists, large and expensive programs, endless conferences. . . and public relations activities is founded on hunch, not evidence, and not on science. . . Contemporary treatment for alcoholism owes its existence more to historical processes than to science....
To determine whether a treatment accomplishes anything we have to know how similar patients who have not received the treatment fare. Perhaps untreated patients do just as well. This would mean that the treatment does not influence outcome at all....
After all [many feel], we have provided many of our treatments for years. We really are confident that the treatment approaches are sound.... Yet, the history of medicine demonstrates repeatedly that unevaluated treatment, no matter how compassionately administered, is frequently useless and wasteful and sometimes dangerous or harmful'.
Miller and Hester , summarized the results of their examination of controlled experimentation with alcoholism treatment:
As we constructed a list of treatment approaches most clearly supported as effective, based on current research, it was apparent that they all had one thing in common. . .: they were very rarely used in American treatment programs. The list of elements that are typically included in alcoholism treatment in the United States likewise evidenced a commonality: virtually all of them lacked adequate scientific evidence of effectiveness'.
Beneficial therapies included aversion therapies, behavioral self-control training, the community reinforcement approach, marital and family therapy, social skills training, and stress management. The standard programs that have not demonstrated their efficacy include Alcoholics Anonymous (A.A.), alcohol education, confrontation, disulfiram, group therapy, and individual counseling. For example, the two controlled experiments involving random assignment to A.A. versus other treatments (or no treatment) favoured the treatments other than A.A. or no treatment [4,5].
The importance of employing comparison groups for developing conclusions about treatment is evident in a long-term outcome study conducted by Vaillant . He carefully tracked his patients in an A.A.-based hospital program for 8 years after treatment and compared their outcomes with those for comparably severe alcoholics in several community studies who had received no treatment. Vaillant reported the following startling disconfirmation of his subjective impressions of the success of his program:
'It seemed perfectly clear that. . . by inexorably moving patients from dependence upon the general hospital into the treatment system of AA, I was working for the most exciting alcohol program in the world. But then came the rub. Fueled by our enthusiasm, I and the director. . . tried to prove our efficacy. Our clinic followed up our first 100 detoxification patients. . . [and found] compelling evidence that the results of our treatment were no better than the natural history of the disease.'
Helzer et al.  found an extremely low remission rate for patients undergoing alcoholism treatment in a highly publicized study. Although the investigators reported as their major conclusion that few patients became moderate drinkers, they did not actually examine controlled-drinking therapy. Rather, they measured outcomes for alcoholics treated in four hospital settings, including an ordinary medical/surgical ward. Of the four settings, the alcoholism treatment ward had the lowest remission rate; only 7% in this group survived and were judged to be in remission at follow-up of from 5 to 7 years. This remission rate is actually substantially lower than that Vaillant noted for a number of untreated groups of alcoholics followed over comparable time periods.
Although Miller and Hester  suggest the possibility that hospital treatment "may be differentially beneficial for more severely deteriorated and less socially stable individuals," their review of all research in which inpatient and other treatments were compared concluded that more intensive treatments have not been found to offer greater benefits than those from outpatient therapy for any population. Indeed, the differences that have been discovered, for example, the likelihood the patient will be re-hospitalized following treatment, favour the outpatient setting. Obviously, if expensive inpatient treatment is ineffective, or no more effective than outpatient counseling or even no treatment, then it will certainly not be judged cost effective and deserving of reimbursement.
Despite data showing hospital treatment is no more efficacious than outpatient counseling, the U.S. Congressional Office of Technology Assessment  reported, 'reimbursement systems. . . have overwhelmingly emphasized the most expensive treatment services inpatient, medically based treatment'. The situation in Canada and Britain is quite different. According to Murray :
'There can be no doubt that current British and American perspectives on alcoholism differ widely . . . British clinicians have shown that the effect of treatment is only marginal, and, in contrast to their American counterparts, have decided against a major expansion of inpatient treatment.... It is perhaps worth noting that whether or not alcoholism is considered a disease, and how much treatment is offered, has no bearing on the remuneration of British doctors'.
In the United States, inpatient care for alcoholism and drug abuse has been undergoing a boom ever since the mid-1970s. The National Institute on Drug Abuse reports that the number of drug treatment centers rose from 3018 in 1982 to 5360 in 1987; between 1978 and 1984, the number of private alcoholism treatment centers quadrupled and the patients treated in them quintupled . These dramatic increases in addiction treatment occurred during a period when illicit drug use and alcohol consumption actually decreased, and such treatment continues to accelerate. According to Health Care Competition Week (July 24, 1989), 'psychiatric, chemical dependency and rehabilitative hospital care, all largely unregulated by government payment mechanisms, are booming'.
An expensive private treatment system in the United States encourages and rewards expensive treatments and has no mechanisms in place to evaluate outcomes. This system is very closely aligned with a powerful "recovering alcoholics" lobby and builds on the naive American acceptance of the efficacy of Alcoholics Anonymous and medical treatment. John Wallace , clinical director of the Edgehill Newport hospital, issued a rallying cry to alcoholism counselors to fight those who question whether intensive alcoholism treatment is cost beneficial:
'These forces of disunity tried first to divide the alcoholism field over the issue of controlled drinking and then, through various attacks upon sobriety, on the disease model of alcoholism, on recovered people, on the concepts, principles, and activities of Alcoholics Anonymous. Now it appears that the target has become the still emerging and fragile comprehensive system of alcoholism treatment services....
We must recognize and resist the various tactics and strategies of the anti-traditionalist lobby to divide us. We must stand shoulder to shoulder in solidarity. Otherwise, alone and divided we will be weak and easy targets for those who do not want to pay for alcoholism services. The most costly outcomes of the current debate over the cost-effectiveness of alcoholism treatment would be the blind and mindless destruction of the comprehensive system of treatment services that benefits so many desperately ill people and took so many years of struggle to build. We cannot, must not let this happen'.
Wallace authored the treatment section of the Sixth Special Report to the U.S. Congress on Alcohol and Health . This report was originally written by Peter Nathan (Director of the Rutgers Center of Alcohol Studies), Barbara McCrady (Clinical Director at the Rutgers Center), and Richard Longabaugh (Director of Evaluation at Butler Hospital in Providence, RI). Nathan et al.'s draft indicated that inpatient treatment produced no greater benefits than did outpatient treatment and that intensive alcoholism treatment was not cost effective. The NIAAA submitted the draft for review by Wallace, who revised it, after which Nathan, McCrady and Longabaugh withdrew their names from the document .
Wallace's argument is that, while comparative evaluations of hospital treatment have shown disappointing results (like those presented by Vaillant and Helzer et al.), this is only because the treatments involved were poor ones. Instead, he points to an outcome study he conducted at his Edgehill Newport Hospital which found 66% of patients were continuously abstinent at 6 months following treatment to show that appropriate treatments can lead to tremendously high remission rates. Wallace  characterizes his study as methodologically rigorous and representative in which "patients were randomly selected from a pool of socially stable patients". Longabaugh  put the study's methodology in a different light, however, in the following analysis:
'[Wallace's] program report was limited to treatment of socially stable patients who were judged to have restorative potential; they had been transferred from detox to rehabilitation indicating that it was expected they would participate fully in a rehabilitation program; they were married and living with a spouse with no plans to separate; they had sufficient resources to pay for treatment; they had asked to participate in the study in the third week of treatment, after any dropouts would have been removed from the sample; they had been 'regularly discharged from the program' with no accounting of patients who were not 'regularly' discharged.'
Was this population representative of the population they were treating? We don't know the answer.... More important, this treatment for this group is not compared with any alternative. It is not compared with a hospital program, an outpatient program, with AA, or no treatment whatsoever.... Any other intervention (might be as effective with such a group], perhaps even including no intervention at all....'.
'What can we generalize?' Longabaugh asked. He said we can't generalize about for-profit, free-standing programs with better-prognosis patients because there have been no results based on controlled comparison research reported to date for those kind of treatment programs. Are such evaluations likely in the near future? He said no applications had been received by the NIAAA for research studies in that area. All that can be expected are single-program studies of doubtful value for such purposes.
On a nationally televised program, Nightline, Wallace made even greater claims for his research :
'There are other intensive inpatient programs like Edgehill Newport that show a dramatically higher recovery rate [than hospitals used in comparison studies]. In our latest randomly assigned study of socially stable alcoholics treated in a middle-class alcoholism treatment program, 66 percent of our people are continuously abstinent from both alcohol and drugs. . . at six months following treatment'.
Here, unfortunately, we have a cheapening of the research dialogue, where Wallace has redefined the term 'randomly assigned' from its usual meaning of assignment to independent treatment groups in a clinical trial to apply to his highly selected group of patients from his unmatched private hospital treatment population. Ultimately some groups may be shown to be helped by inpatient treatment as part of a panoply of alcoholism services. The target population for hospital treatment, however, is unlikely to be the socially stable, middle-class alcoholics included in the Edgehill Newport study, who are those most likely to overcome drinking problems under any regimen or even without treatment.
In the meantime, the failure to subject treatment approaches to systematic evaluation will not benefit alcoholics in the way advocates of private hospital treatment seem to hope, but will only make it harder to discover which treatments are best for which patients. Only when private treatment centers are motivated to participate in actual clinical trials will an effective alcoholism treatment system be possible in the United States.
- W.R. Miller and R.K. Hester, Am. Psychol., 41 (1986) 794.
- E. Gordis, J. Stud. Alcohol,48 (1987) 579.
- W.R. Miller and R.K. Hester, The effectiveness of alcoholism treatment: What the research shows, in: W.R. Miller and N.K. Heather (Eds.). Treating Addictive Behaviors, Plenum, New York, 1986, pp. 121173.
- J.M. Brandsma, M.C. Maultsby and R.J. Welsh, The Outpatient Treatment of Alcoholism, University Park Press, Baltimore, MD, 1980.
- K.S. Ditman et al., Am. J. Psychiatry, 124 (1967) 160.
- G.E. Vaillant, The Natural History of Alcoholism, Harvard University Press, Cambridge, MA, 1983, p.283.
- J.E. Helzer et al., N. Engl. J. Med.,312 (1985) 1678.
- U.S. Congress, Office of Technology Assessment, The Effectiveness and Costs of Alcoholism Treatment, p.66. Government Printing Office, Washington, DC, 1983.
- R.M. Murray et al., Economics, occupation and genes: A British perspective, paper presented to the American Psychopathological Association, New York, March, 1986.
- R. Longabaugh, Optimizing the cost effectiveness of treatment, in: Evaluation Recovery Outcomes, University of California at San Diego Extension Program on Alcohol Issues, 1988, pp.22.
- J. Wallace, J. Profess. Counselor May/June, 1987,23.
- J. Wallace, Chapter VII, Treatment, Sixth Special Report to the U.S. Congress on Alcohol and Health from the Secretary of Health and Human Services, DHHS Publication, Rockville, MD, 1987.
- W.R. Miller, Adv. Behav. Res. Ther., 9 (1987) 154.
- J. Wallace, J. Psychoactive Drugs, 21 (1989) 259.
- Nightline, Alcoholism Treatment Controversy, ABC News, New York, 1989.